TCONS_00230836 silencing restores stearic acid- induced β-cell dysfunction through alleviating endoplasmic reticulum stress via a PERK/eIF2α- dependent pathway rather than apoptosis


 Background: Chronic exposure of pancreatic β cells to high levels of stearic acid (C18:0) leads to impaired insulin secretion, which accelerates the progression of type 2 diabetes mellitus (T2DM). Recently, long noncoding RNAs (lncRNAs) were found to participate in saturated fatty acid-induced metabolism dysfunction. However, their contribution to stearic acid-induced β-cell dysfunction remains largely unknown. This study evaluated the possible role of the lncRNA TCONS_00230836 in stearic acid-stimulated lipotoxicity to β cells. Method: Using high-throughput RNA-sequencing, TCONS_00230836 was screened out as being exclusively differentially expressed in stearic acid-treated mouse β-TC6 cells. Co-expression network was constructed to reveal the potential mRNAs targeted for lncRNA TCONS_00230836. Changes in this lncRNA’s and candidate mRNAs’levels were further assessed by real-time PCR in stearic acid-treated β-TC6 cells and islets of mice fed a high-stearic-acid diet (HSD). The localization of TCONS_00230836 was detected by fluorescent in situ hybridization. The endogenous lncRNA TCONS_00230836 in β-TC6 cells was abrogated by its Smart Silencer. Results: The lncRNA TCONS_00230836 was enriched in mouse islets and mainly localized in the cytoplasm. Its expression was significantly increased in stearic acid-treated β-TC6 cells and HSD-fed mouse islets. Knockdown of TCONS_00230836 apparently restored stearic acid-impaired GSIS through alleviating endoplasmic reticulum stress via a PERK/eIF2α-dependent pathway. However, stearic acid-induced β-cell apoptosis was not obviously recovered. Conclusion: Our findings suggest the involvement of the lncRNA TCONS_00230836 in stearic acid-induced β-cell dysfunction, which provides novel insight into stearic acid-induced lipotoxicity to β cells. Anti-lncRNA TCONS_00230836 might be a new therapeutic strategy for alleviating stearic acid-induced β-cell dysfunction in the progression of T2DM.


Introduction
Chronic exposure to elevated saturated fatty acids (SFAs) of pancreatic β cells is a key trigger of impaired insulin secretion, which is one of the most important characteristics of type 2 diabetes mellitus (T2DM).
Evidence accumulated to date has mainly focused on palmitic acid, with little attention being paid to stearic acid-induced worsening of β-cell function. In our previous studies and other research, it has been demonstrated that, whether in the postprandial serum of T2DM patients or in the fasting serum of highfat-diet-fed mice [1][2][3][4], only stearic acid levels were profoundly increased. Meanwhile, a long-term high level of stearic acid exhibited a stronger destructive effect on β cells than that of palmitic acid [2,5]. Although endoplasmic reticulum (ER) stress [2,6] and apoptosis [5,7] are accepted as major contributors to stearic acid-induced β-cell dysfunction, the molecular mechanisms involved in this remain largely unclear.
Twenty male C57BL/6J mice (6 weeks old) were purchased from Beijing Vital River Laboratory Animal Technology Company (Beijing, China). After adaptive feeding for 1 week, these mice were randomly divided into control and high-stearic-acid diet groups (n = 10 per group), according to their body weights. The control diet (1025) and high-stearic-acid diet (HSD) (H10060) were obtained from Beijing HFK Bioscience Co. Ltd. (Beijing, China) (Additional le 1). After 20 weeks of feeding the mice, islets and blood samples were collected for biochemical analysis. Fasting (12 hours) serum glucose, total cholesterol, triacylglycerol, high-density-lipoprotein cholesterol, and low-density-lipoprotein cholesterol levels were calculated using an automatic analyzer (Hitachi-7100; Hitachi, Tokyo, Japan), kits for which were purchased from Biosino Biotechnology Co. (Beijing, China). Serum insulin level was measured using a mouse/rat insulin ELISA kit (Linco Research, St. Charles, MO, USA).
Serum nonesteri ed fatty acid pro le measurement.
Fasting serum nonesteri ed fatty acids were transformed to fatty acid methyl esters, as described in our previous studies [4,20]. Gas chromatography-mass spectrometry analysis was performed using a TRACE gas chromatograph with a Polaris Q mass spectrometer (Thermo Finnigan, San Jose, CA, USA). Separation was obtained on a J&W DB-WAX capillary column (30-m, 0.25-mm I.D., 0.25-µm lm thickness; Agilent J&W Scienti c, Folsom, CA, USA). Heptadecanoic acid (C17:0) was used as an internal standard.
Cell viability measurements.
Cell viability was measured by the assessment of lactate dehydrogenase (LDH) release and CCK 8 assay. For measuring LDH, the culture medium was collected and characterized using LDH assay kit (Thermo Fisher Scienti c Inc.). For the CCK 8 assay, we used the Cell Counting Kit 8 (CCK-8, C0038; Beyotime Biotechnology, Shanghai, China). The β-TC6 cells were seeded into each well of a 96-well plate and 100 µL of CCK-8 mixed reagents were added to each well. After 1 h of incubation at 37 °C, absorbance was determined at 450 nm using a microplate reader (SpectraMax M2; Molecular Devices, San Jose, CA, USA). Apoptosis assay.
The β-TC6 cells were seeded into a 24-well plate with sterile slides overnight. Cells were rinsed using 1 × PBS for 5 min and xed in 4% formaldehyde at room temperature for 10 min. The β-TC6 cells were washed three times with 1 × PBS for 5 min each. Next, the cells were permeabilized in 1 × PBS containing 0.5% Triton X-100 for 5 min at 4 °C. After washing the cells with 1 × PBS, they were blocked in Blocking Solution and Pre-hybridization (1:99) mixed solution at 37 °C for half an hour. Then, in the dark, discarding the mixed solution, the β-TC6 cells were incubated with a hybridized mixture containing lncRNA Probe Mix or U6/18S at 37 °C overnight. On the second day, the β-TC6 cells were washed at 42 °C with different concentrations of SSC solution, which were also protected from the light. After immersing the β-TC6 cells in 1 × PBS for 5 min, they were stained with DAPI staining solution at room temperature for 10 min and then washed with 1 × PBS. The slides taken from the 24-well plate were observed with a laser confocal microscope. The whole process was protected from light to prevent quenching. The locked nucleic acid-modi ed oligonucleotide probe targeting lncRNA TCONS_00230836 and the FISH Kit were purchased from Ribobio Co. Ltd. (Guangzhou, China).

Transfection procedures.
Smart Silencer for TCONS_00230836 and its negative control (Ribobio Co. Ltd.) were transfected into β-TC6 cells using C10511-05 riboFect™ CP Transfection Kit (Ribobio Co. Ltd.) for 24 h, in accordance with the manufacturer's instructions. The sequences of lncRNA Smart Silencer for mouse TCONS_00230836 are displayed in Table 1, including three siRNA and three antisense oligonucleotides. Quantitative real-time polymerase chain reaction (qRT-PCR).
Total RNA was extracted from β-TC6 cells and mouse islets using TRIzol reagent (Invitrogen), in accordance with the manufacturer's protocol. The qRT-PCR was performed with SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA), with β-actin as an internal control. All primers used in this study were synthesized by Sangon Biotech Co. Ltd. (Shanghai, China), the sequences of which are listed in Table 2.  Statistics.
All statistical data were analyzed with SPSS software, version 21.0 (SPSS Inc., Chicago, IL, USA), and are reported as mean ± SEM. Two-tailed Student's t-test was used to analyze differences between two groups, and one-way ANOVA followed by Student-Newman-Keuls test was used to test differences among three or more groups. P-values < 0.05 were considered statistically signi cant.

Results
Animal model characteristics.
We successfully established a mouse model to mimic the increased proportions of stearic acid in vivo, as shown by a signi cant increase in circulating stearic acid in HSD-fed mice (Additional le 2). The metabolic characteristics of these animals are displayed in Additional le 3.
Upregulation of lncRNA TCONS_00230836 expression in stearic acid-treated β-TC6 cells and islets of mice fed a high-stearic-acid diet.
The results obtained by high-throughput RNA-sequencing demonstrated that the lncRNA TCONS_00230836, an intergenic lncRNA located at 65803111 to 65809116 of chromosome 10, was markedly elevated with a 1.328 log 2 fold change (P value: 5.062E-07) in β-TC6 cells exposed to stearic acid and a 0.270 log 2 fold change (P value: 0.327) in the palmitic group, compared with the level in the normal group (Fig. 1a). A similar increase in lncRNA TCONS_00230836 level was also revealed in both stearic acid-treated β-TC6 cells and islets of mice fed a high-stearic-acid diet by qRT-PCR. The expression level of the lncRNA TCONS_00230836 was increased by 3.744-fold in stearic acid-treated β-TC6 cells (Fig. 1b) and 5.115-fold in the islets of mice from the high-stearic-acid group (Fig. 1c), compared with that in the control group.
Differential expression of lncRNA TCONS_00230836 in various tissues and cell lines of mouse.
qRT-PCR results veri ed that the level of the lncRNA TCONS_00230836 was signi cantly differentially expressed in various tissues as well as cell lines those are closely related to high-fat-diet induced metabolic disorders. Among them, the expression of this lncRNA was highest in the islets and lowest in the skeletal muscle of HSD-fed mice (Fig. 2a). However, no matching sequence was detected in peritesticular and perirenal adipose tissues (data not shown). Meanwhile, the expression of the lncRNA TCONS_00230836 was apparently higher in β-TC6 cells than that in AML12 cells (Fig. 2b). In addition to islets and β-TC6 cells, after the treatment with a high concentration of stearic acid, the lncRNA TCONS_00230836 level was also increased in brown adipose, whereas it was decreased in liver and skeletal muscle of HSD-fed mice and AML12 cells (Additional le 4).
lncRNA TCONS_00230836 is mainly localized in cytoplasm.
The confocal images of FISH showed that the lncRNA TCONS_00230836 is mainly distributed in the cytoplasm of β-TC6 cells (Fig. 3).
Both in islets and β-TC6 cells ( Fig. 4a and 4b), transfection of the lncRNA TCONS_00230836 Smart Silencer for 24 h e ciently reduced the stearic acid-increased intracellular lncRNA TCONS_00230836 levels. Meanwhile, similar results were consistently observed in β-TC6 cells as indicated by FISH (Fig. 4c).
Then, as shown in Fig. 4d and 4e, stearic acid markedly decreased GSIS in islets and β-TC6 cells, separately. After transfection of the lncRNA TCONS_00230836 Smart Silencer into pancreatic islets and β-TC6 cells, stearic acid-induced GSIS impairment was signi cantly alleviated. However, no signi cant alteration was observed in GSIS when the lncRNA TCONS_00230836 Smart Silencer was transfected alone in the absence of stearic acid.
Inhibition of lncRNA TCONS_00230836 alleviated stearic acid-induced ER stress rather than apoptosis in β-TC6 cells and mice islets.
Stearic acid apparently suppressed the cell survival rate and induced cell death in β-TC6 cells. Silencing the lncRNA TCONS_00230836 did not signi cantly block the stearic acid-stimulated cell viability reduction (Fig. 5a) and cell death (Fig. 5b). Meanwhile, the annexin V-FITC apoptosis assay results showed that lncRNA TCONS_00230836 inhibition did not reduce the stearic acid-increased percentage of apoptotic cells (Fig. 5c). Besides, the elevated protein levels of cleaved-Parp1 and cleaved-Caspase3, and the decreased levels of B-cell CLL/lymphoma 2 in the stearic acid group were not rescued by the lncRNA TCONS_00230836 Smart Silencer (Fig. 5d). However, the expression of the ER stress-related proteins Phospho-PERK and Phospho-eIF2α in stearic acid-treated β-TC6 cells was signi cantly decreased after inhibition of the lncRNA TCONS_00230836 (Fig. 6a), whereas stearic acid-increased protein levels of ATF-6 and IRE1α were not suppressed after knockdown of TCONS_00230836 (Fig. 6b). Moreover, transfection of the lncRNA TCONS_00230836 Smart Silencer alone did not alter apoptosis and ER stress in β-TC6 cells. In addition, similar results were consistently observed in mice islets (Fig. 7a-c, Fig. 8a, b).
Effect of lncRNA TCONS_00230836 on the expressions of candidate ER stress-related mRNAs.
In order to reveal the potential mRNAs targeted for lncRNA TCONS_00230836 in stearic acid-treated β-cell ER stress, the lncRNA TCONS_00230836-mRNAs co-expression network was constructed. 12 up-regulated mRNAs are positively associated with lncRNA TCONS_00230836 (Fig. 9a), and 15 down-regulated mRNAs are negatively related to this lncRNA (Fig. 9b) were signi cantly up-regulated after silencing lncRNA TCONS_00230836 in the absence of stearic acid (Fig. 9d). The log2 fold change of these top ve up-and down-regulated differentially expressed mRNAs were listed in Additional le 5.

Discussion
The prevalence of T2DM has been increasing worldwide, with an estimated 336 million people currently affected [21]. It is well established that exposing pancreatic β cells to an elevated level of stearic acid causes their dysfunction, which is far advanced by the time diabetes is diagnosed clinically [1][2][3]5]. Therefore, the optimal management and prevention of T2DM should aim to ameliorate β-cell dysfunction at an early stage. However, the effective preventive targets are limited and still need to be comprehensively explored. The present study elucidated the pathophysiological role of lncRNA in stearic acid-treated β cells. Our ndings demonstrated that the lncRNA TCONS_00230836 was signi cantly increased in both stearic acid-induced β-TC6 cells and the islets of mice fed a high-stearic-acid diet.
Inhibition of the lncRNA TCONS_00230836 effectively protected against stearic acid-induced β-cell dysfunction. Interestingly, after knocking down the lncRNA TCONS_00230836, we found that stearic acidstimulated β-cell ER stress was apparently ameliorated via a PERK/eIF2α-dependent pathway, but no obvious improvement of stearic acid-induced apoptosis was observed.
Although a number of studies have preliminarily explored the mechanisms underlying stearic acidinduced pancreatic β-cell damage, the exact molecular targets and associated pathways still need to be established. Recently, accumulating evidence has suggested that noncoding RNAs play an important role in the β-cell injury caused by the long-term consumption of a high-fat diet, such as microRNAs [2,5], circRNAs [22,23], and lncRNAs [17,24,25]. Nevertheless, few studies have focused on the role of lncRNAs in β-cell dysfunction upon exposure to elevated levels of SFAs, especially stearic acid, on which no related research has previously been published. In our study, we used high-throughput sequencing and bioinformatic technology to select ve lncRNAs that were differentially expressed speci cally in stearic acid-treated β-TC6 cells, compared with their levels in both control and palmitic acid groups (data not shown). Among the known lncRNAs, the lncRNA TCONS_00230836 displayed the largest fold increase, with an increase of 1.328 log 2 fold change in RNA sequencing results, 3.744-fold in stearic acid-induced β-TC6 cells, and 5.115-fold in islets of mice fed a high-stearic-acid diet, as revealed by qRT-PCR.
It has been well clari ed that lncRNAs have low sequence conservation and high tissue speci city [26]. Meanwhile, the expression of lncRNAs typically varies more between tissues than for other noncoding RNAs [27][28][29]. To investigate the difference of sequence conservation and differential expression of the lncRNA TCONS_00230836, we tested it in different tissues (islet, liver, skeletal muscle, brown adipose, peritesticular, and perirenal adipose tissues) and cell lines (β-TC6 cells and AML12 cells) which are closely related to T2DM. Our data indicated that there was a highly similar sequence of lncRNA TCONS_00230836 (similarity > 98%) in liver, brown adipose, skeletal muscle, as well as the AML12 cell line, but not in peritesticular and perirenal adipose tissues, compared with that in islets and β-TC6 cells. Moreover, the level of this lncRNA was highest in islets and lowest in skeletal muscle of high-stearic-acidfed mice, which suggested that this lncRNA may be speci c to and enriched in islets. This is the rst key nding of our study.
To demonstrate the role of the lncRNA TCONS_00230836 in stearic acid-mediated lipotoxicity in β cells, endogenous TCONS_00230836 in β-TC6 cells was abrogated by the application of its Smart Silencer, a highly effective inhibitor that speci cally targets lncRNA TCONS_00230836 expressed in both the cytoplasm and the nucleus. We found that stearic acid-impaired GSIS was largely restored after silencing TCONS_00230836, which implied that a TCONS_00230836-mediated mechanism operated in the stearic acid-induced lipotoxicity of β cells. However, how does the lncRNA TCONS_00230836 exert its in uence and which process was improved in stearic acid-induced pancreatic β-cell injury?
ER stress and apoptosis are very important pathophysiological perturbations and are closely associated with β-cell lipotoxicity in the early stage of T2DM; they are not only independent of each other, but also have a close cause-effect relationship [30][31][32][33][34]. To reveal which process was affected by the lncRNA TCONS_00230836, we detected the alterations of ER stress and apoptosis induced by stearic acid after transfection of the Smart Silencer for the lncRNA TCONS_00230836. Our data indicated that transfection of the lncRNA TCONS_00230836 effectively ameliorated stearic acid-stimulated ER stress via a PERK/eIF2α-dependent pathway. As we know, ER stress can transduce apoptotic signals that may eventually result in apoptosis. However, unexpectedly, stearic acid-induced β-cell apoptosis was not signi cantly recovered after knocking down lncRNA TCONS_00230836 expression. Therefore, we speculate that there may be some other key regulatory factors involved in lncRNA TCONS_00230836mediated β-cell ER stress. This is another important nding of our study.
In an effort to reveal the potential target for lncRNA TCONS_00230836 in stearic acid-induced ER stress, we analyzed the mRNAs pro le by high-throughput screens in our previous study [35] and constructed the co-expression network between lncRNA TCONS_00230836 and targeted mRNAs which are closely related to ER stress. Then top ve up-regulated mRNAs and another top ve down-regulated mRNAs were selected for further identi cation. In the positive-relation group, qPCR results for the mRNA expressions indicated that Alpk1, Icam1 and Serping1 are probably involved in the process of stearic acid-induced βcell ER stress associated with lncRNA TCONS_00230836. While in negative-relation group, Prn, Bcas3, Otub2, Afmid and Srgap2 might play a critical role in stearic acid-β-cell ER stress closely related to lncRNA TCONS_00230836. Among them, Afmid has been reported that they could lead to impaired glucose tolerance in type 2 diabetes [36]. Meanwhile, Otub2 has been shown to act through the inhibition of NF-κB signaling in type 1 diabetes [37,38]. Further experiments should be performed to explore the precise regulatory relationship between these candidate mRNAs and lncRNA TCONS_00230836 as well as PERK/eIF2α-dependent signaling in stearic acid-treated β cells.
Our understanding of lncRNAs in stearic acid-induced lipotoxicity to β cells is still in its infancy and several questions remain unanswered. First, we did not assess the role of the lncRNA TCONS_00230836 in stearic acid-induced impairment of insulin secretion in an in vivo experiment. Second, we all know that lncRNAs exert their function through diverse regulatory mechanisms, for example, modulating translation, promoting mRNA degradation, or acting as miRNA sponges. Therefore, further experiments still need to be performed to improve our understanding of the regulatory mechanisms of the lncRNA TCONS_00230836 in stearic acid-induced lipotoxicity to pancreatic β cells.

Conclusions
We provide evidence that stearic acid can stimulate upregulation of the lncRNA TCONS_00230836 in stearic acid-treated pancreatic β cells. It appears that this lncRNA is speci c to and enriched in pancreatic

Consent for publication
Not applicable.
Availability of data and material All of the data are available with reasonable request from the corresponding authors.

Competing interests
The authors declare that there is no con ict of interest. level of the lncRNA TCONS_00230836 was elevated in stearic acid-treated β-TC6 cells. (n=4) (c) The expression of the lncRNA TCONS_00230836 was also increased in islets from mice fed a high-stearicacid diet, as revealed by qRT-PCR. Ctrl, control group; SA, stearic acid; HSD, high-stearic-acid diet. **P < 0.01, versus the Ctrl group.  Confocal images of FISH for localization of the lncRNA TCONS_00230836 (red) in β-TC6 cells. Nuclei were stained with DAPI (blue). 18S, probe for 18S rRNA, was used for cytoplasmic localization. U6, probe for U6 snRNA, was taken as a representative of nuclear localization. Scale bar, 10 μm.     and IRE1α protein levels after transfection of the TCONS_00230836 Smart Silencer in islet exposed to high level of stearic-acid. NS, not signi cant; SA, stearic-acid; si-lnc836, transfection of the Smart Silencer for the lncRNA TCONS_00230836. *P < 0.05 versus the control group; ^P < 0.05 versus SA group. n = 5 independent animal experiments per group.